v23.5 seems to have activated speedstepping in my notebook?!

E_tron · 2003-08-08, 16:07

Ok very strange indeed, but actually worked to my advantage :D . v23.5 seems to have activated speed stepping in my notebook

.

Before v23.5 (according to cpuz and prime95) my P3-M(low power version) ran at 133x7.5=1000mhz all the time, idle or not. Now, after running v23.5 and v23.6 the cpu goes to 133x5.5=733mhz(ish) when the cpu is idle or not exceeding 80% cpu utilization(@733). But, when i kick up the load (like p95 or RTCW:ET) it bumps back to 1ghz.

Has anyone else experienced this type of strange energy saving event before :D ?

Con:
its not really 1ghz anymore :)
pro:
battery life has almost doubled(when not gaming) :D . I never thought that p95 could actually increase battery life.

Xyzzy · 2003-08-08, 16:10

I thought SpeedStep changed the FSB, not the multiplier...

IOW, you might be going from 7.5 x 133 to 7.5 x 100...

Xyzzy · 2003-08-08, 16:22

Looks like I am wrong...

Quote:

When powered by a battery, the processor drops to a lower frequency (by changing the bus ratios) and voltage, conserving battery life while maintaining a high level of performance.

http://support.intel.com/support/processors/mobile/pentiumiii/ss.htm

My P3-M will only go to low-power mode if I use the battery... I've tried to force it there but I run Linux and I think I have to do it by hand or something... My low-power setting is from 1200MHz to 800MHz...

E_tron · 2003-08-08, 16:23

cpuz says it is the multiplier. I would think that it would be easier on the system to just change the multiplier and not the entire system bus.

Isn’t the multiplier unchangeable on most intels?

E_tron · 2003-08-08, 16:27

i wonder if i can force it lower?. the cpu is about the only thing other than the wifi that uses up energy.

Xyzzy · 2003-08-08, 16:27

Quote:

Originally Posted by E_tron

Isn’t the multiplier unchangeable on most intels?

On the desktop CPUs only the "confidential" sample/engineering chips are unlocked... I never paid much attention to the mobile chips before...

Xyzzy · 2003-08-08, 16:35

Quote:

Originally Posted by E_tron

i wonder if i can force it lower?. the cpu is about the only thing other than the wifi that uses up energy.

Any lower and you would be better off buying a Mac...

:)

The chart in that Intel link says 700MHz is the only option, but who knows?

Here is the full documentation...

ftp://download.intel.com/design/mobile/datashts/p3_ds.pdf

Look in the electrical section... It will list which pins have to be set to adjust this... (Assuming it uses pin combos to do that... A lot of stuff, like voltage selection, are determined by "shorting" certain pin combos together... In the "old days" that was how I used to overvolt CPUs... I'd take hair-thin wire and wrap it around the pins before I put the CPU in the socket!)

E_tron · 2003-08-08, 16:40

This has turned out to be a hardware discussion :D . Wrong forum :D .

QuintLeo · 2003-08-09, 03:28

*looks around for a software forum moderator to move this topic*

9-)

aaronl · 2003-08-19, 08:59

Yeah, I had this problem too. My laptop is pretty lame in the sense that it's built around a desktop P4 processor. I never used to get the speedstepping, but now that it's summer and/or the cooling devices are breaking down, I can't run mprime for more than a few minutes without the cpu speed more than halving. I've been working on this one exponent since May with limited hours whenever the laptop is in use. And the laptop quickly gets too hot to be placed on a lap.

2003-08-08, 16:07	#1
E_tron Sep 2002 Austin, TX 561₁₀ Posts	v23.5 seems to have activated speedstepping in my notebook?! Ok very strange indeed, but actually worked to my advantage :D . v23.5 seems to have activated speed stepping in my notebook . Before v23.5 (according to cpuz and prime95) my P3-M(low power version) ran at 133x7.5=1000mhz all the time, idle or not. Now, after running v23.5 and v23.6 the cpu goes to 133x5.5=733mhz(ish) when the cpu is idle or not exceeding 80% cpu utilization(@733). But, when i kick up the load (like p95 or RTCW:ET) it bumps back to 1ghz. Has anyone else experienced this type of strange energy saving event before :D ? Con: its not really 1ghz anymore :) pro: battery life has almost doubled(when not gaming) :D . I never thought that p95 could actually increase battery life.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Notebook fan cleaning in song	kladner	Hardware	2	2013-06-01 06:00
electronic notebook paper	jasong	jasong	14	2012-01-23 03:32
Ancient notebook won't run Prime95	kwstone	Hardware	4	2004-01-15 00:11

2003-08-08, 16:10	#2
Xyzzy Aug 2002 2×19×223 Posts	I thought SpeedStep changed the FSB, not the multiplier... IOW, you might be going from 7.5 x 133 to 7.5 x 100...

2003-08-08, 16:23	#4
E_tron Sep 2002 Austin, TX 3×11×17 Posts	cpuz says it is the multiplier. I would think that it would be easier on the system to just change the multiplier and not the entire system bus. Isn’t the multiplier unchangeable on most intels?

2003-08-08, 16:27	#5
E_tron Sep 2002 Austin, TX 3·11·17 Posts	i wonder if i can force it lower?. the cpu is about the only thing other than the wifi that uses up energy.

2003-08-08, 16:40	#8
E_tron Sep 2002 Austin, TX 3·11·17 Posts	This has turned out to be a hardware discussion :D . Wrong forum :D .

2003-08-09, 03:28	#9
QuintLeo Oct 2002 Lost in the hills of Iowa 2⁶×7 Posts	looks around for a software forum moderator to move this topic 9-)

2003-08-19, 08:59	#10
aaronl Aug 2003 2⁴×3 Posts	Yeah, I had this problem too. My laptop is pretty lame in the sense that it's built around a desktop P4 processor. I never used to get the speedstepping, but now that it's summer and/or the cooling devices are breaking down, I can't run mprime for more than a few minutes without the cpu speed more than halving. I've been working on this one exponent since May with limited hours whenever the laptop is in use. And the laptop quickly gets too hot to be placed on a lap.